httpd-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jack Levin" <cali...@calibri.net>
Subject [users@httpd] web farm cluster (NFS/Apache).
Date Fri, 02 Apr 2004 21:41:47 GMT
Greetings,

I have 6 x Redhat 9, linux machines, where 1 is NFS "master" while
the other 5 are NFS slaves.   My domain name is a list of A records
for all 6 machines (IPs), allowing for round robin traffic load
distribution.
All machines run apache, and do read/write to the NFS shared directory.

Everything runs great, except that when machine 1 (NFS master) goes
down for reload or NFS share otherwise is no longer available -
all slaves quickly run out of memory and IOwait goes to 100%, as
bunch of httpd processes pile up with "DN" stat.

I was thinking to mitigate this problem with another NFS server that
I could quickly have slaves switch to, when primary NFS is down.
And then switch back to the primary NFS when things are back
to normal.

Does anyone have any experience doing that?

Any insight would be greatly appreciated!

Thanks,

-Jack


---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
   "   from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org


Mime
View raw message